AWS Academy Machine Learning Foundations

Module 2: Introduction to
Machine Learning

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module overview

Sections Demonstration
Introducing Amazon SageMaker
1. What is machine learning?
2. Business problems solved with
machine learning
3. Machine learning process
4. Machine learning tools overview
5. Machine learning challenges
Knowledge
check

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

2
reserved.
Module objectives

At the end of this module, you should be able to:

• Recognize how machine learning and deep learning are part of artificial
intelligence
• Describe artificial intelligence and machine learning terminology
• Identify how machine learning can be used to solve a business problem
• Describe the machine learning process
• List the tools available to data scientists
• Identify when to use machine learning instead of traditional software
development methods

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

3
reserved.
Module 2: Introduction to Machine Learning

Section 1: What is machine learning ?

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
Artificial intelligence, machine learning, and
deep learning

Artificial intelligence

Machine learning

Deep learning

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

5
reserved.
Artificial intelligence

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

6
reserved.
Machine learning

Machine learning is the scientific study of algorithms and statistical

models to perform a task using inference instead of instructions.

Data Model Prediction

Machine learning flow

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

7
reserved.
Deep learning

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

8
reserved.
ML and technology advancements

Traditional Cloud computing Modern machine

computing and learning
Moore’s law
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
9
reserved.
 • Artificial intelligence
Section 1 key • Machines performing human tasks
takeaways • Machine learning
• Training models to make predictions
• Deep learning
• Neural networks
• Technology and economic
advancements have made
machine learning more
accessible to individuals and
organizations

10 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module 2: Introduction to Machine Learning

Section 2: Business problems solved with

machine learning

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
Common business use cases

Recommended items

Spam versus Recommendations Fraud

regular email

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

12
reserved.
Types of machine learning

Supervise Reinforceme
Machine
d nt
learning
learning learning

Unsupervise
d
learning

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

13
reserved.
Supervised learning
Learn by identifying patterns in data that is already labeled.
Binary [0,1] • Fraud detection
Classificatio • Image recognition
n • Customer retention
Multi [0,1,2] • Medical diagnostics
Supervise
d • Personalized
learning advertising
• Product sales prediction
Regressio • Weather forecasting
n
• Market forecasting
• Population growth prediction
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
14
reserved.
Computer vision

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

15
reserved.
Unsupervised learning

The machine must uncover and create the labels

itself.
• Product recommendations
• Customer segmentation
Clustering
• Targeted marketing
Unsupervise
• Medical diagnostics
d
learning
• Visualization
Dimensionalit
y
• Natural language
reduction processing
• Data structure discovery
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
reserved. • Gene sequencing 16
Natural language processing

gögn eru lykilatriði los datos son

Hi! clave

I’d like two

chickens,
please. Once upon
a time

Right, two
chickens, let me
get that started
for you.

Poor story. Little character development. Jumps between scenes like

you might get caught stealing. Unexplained bad guys appear with the
thinnest of back story. Back to unlimited resources and lets not talk
about the mechanics of building something so huge in such an
inhospitable place...
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
17
reserved.
Reinforcement learning

Learning through trial and error.

• Game AI
Reinforceme
nt
• Self-driving cars
learning • Robotics +1+
1
• Customer service routing
AWS DeepRacer
Best when the desired outcome is known but
the exact path to achieving it is not known.
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
18
reserved.
Self-driving vehicles

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

19
reserved.
When to use machine learning?

Classical programming
Use machine learning when you
approach have:
✔ Large datasets, large number of variables
Business ✔ Lack of clear procedures to obtain the solution
Task
logic
Procedure ✔ Existing machine learning expertise
s
✔ Infrastructure already in place to support ML
✔ Management support for ML
Machine learning
approach

Data Model Task

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

20
reserved.
 • Machine learning applications
Section 2 key affect everyday life
takeaways • Machine learning can be grouped
into –
• Supervised learning
• Unsupervised learning
• Reinforcement learning
• Most problems are supervised
learning

21 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module 2: Introduction to Machine Learning

Section 3: Machine learning process

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
 ML pipeline: Business problem

Business problem

Problem
formulation

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

23
reserved.
 ML pipeline: Data preparation
Data handling and
Business problem
cleaning
data

Problem data data

formulation
data

Collect and Name Country Sex dob

Evaluate data
label data
Richard Roe UK Male 18/2/1972
Paulo Santos Male 11/2/1969
Mrs. Mary Major Denver F 37
Desai, Arnav USA M 2/22/1962

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

24
reserved.
 ML pipeline: Iterative model training

Business problem

Problem
Tune model
formulation

Meets
Collect and Feature Select and Evaluate
Evaluate data busines
label data engineering train model model
s goal?

No
Feature augmentation

Data augmentation

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

25
reserved.
ML pipeline: Feature engineering
Name Country Sex dob
Richard Roe UK Male 18/2/1972
Paulo Santos Male 11/2/1969
Mrs. Mary Major Denver F 37
Desai, Arnav USA M 2/22/1962
?
Name USA UK sex age bm dow target
Richard Roe 0 1 0 49 2 5 140,000
Paulo Santos 1 0 0 51 11 7 78,000
Mary Major 1 0 1 37 NAN 0 167,000
Arnav Desai 1 0 0 58 2 4 100,000

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

26
reserved.
ML pipeline: Model training

Name USA UK sex age bm dow target

10–20%
Richard Roe 0 1 0 49 2 5 140,000 Test data
… … … … … … … …

80%

Algorithm
Trained model
XGBoost

{hyperparameters}

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

27
reserved.
ML pipeline: Evaluating and tuning the
model
Name USA UK sex age bm dow target
10-20%
Richard Roe 0 1 0 49 2 5 140,000 Test data
… … … … … … … …

Change 80% predict

features

Algorithm
Trained model Hosted model
XGBoost

{hyperparameters}
Change hyperparameters

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

Metrics
28
reserved.
Overfitting and underfitting

Y Y Y

X X X
Overfitting Underfitting Balanced

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

29
reserved.
 ML pipeline: Deployment
New data, retraining
Deploy model
Business problem

Problem Yes
Tune model
formulation

Meets
Collect and Feature Select and Evaluate
Evaluate data busines
label data engineering train model model
s goal?

No
Feature augmentation

Data augmentation

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

30
reserved.
 • Machine learning pipeline guides
Section 3 key you through the process of
takeaways evaluating and training a model
• Iterative process of –
• Data processing
• Training
• Evaluation

31 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module 2: Introduction to Machine Learning

Section 4: Machine learning tools overview

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
Python tools and libraries

• Jupyter Notebook • pandas

• JupyterLab • Matplotlib
• Seaborn
• NumPy
• scikit-learn

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

33
reserved.
Machine learning frameworks and
infrastructure
Machine learning frameworks provide Amazon instances that are designed
tools and code libraries: for machine learning applications:
• Customized scripting • AWS IoT Greengrass provides an
• Integration with AWS services infrastructure for building machine
learning for IoT devices
• Community of developers
• Amazon Elastic Inference reduces costs
for running machine learning applications

PyTorch Caffe2 Torch

TensorFlow Gluon Chainer

EC2 P3 EC2 C5 AWS IoT Amazon
instances and C5n Greengrass Elastic
Keras CNTK Apache MXNet
instance Inference
s

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

34
reserved.
 Amazon SageMaker

Ground Truth Notebook Training Inference AWS

Set up and manage Provide AWS and Train and tune Create models from Marketplace
Find, buy, and deploy
labeling jobs for SageMaker SDKs models at any training jobs, or import ready-to-use model
highly accurate and sample scale. Use external models for packages,
training datasets by notebooks to create high-performance hosting so you can run algorithms, and data
using active training jobs and AWS algorithms, inferences on new data. products in AWS
learning and human deploy models. or bring your own. Marketplace.
labeling.

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

35
reserved.
Demonstration:
Introducing
Amazon
SageMaker

36 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Machine learning managed services

These managed services don’t require ML experience.

Compute Chatbots
r
vision
Amazon Rekognition Amazon Textract Amazon
Lex
Speec Forecastin
h g

Amazon Polly Amazon Amazon

Transcribe Forecast
Languag Recommendations
e

Amazon Comprehend Amazon Amazon

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights Translate Personalize 37
reserved.
 • Python is the most popular ML
Section 4 key language
takeaways • Jupyter Notebooks
• Many open-source tools
• Frameworks and services for all
requirements
• Low-level frameworks
• Amazon SageMaker
• Managed ML services

38 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module 2: Introduction to Machine Learning

Section 5: Machine learning challenges

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
 Machine learning challenges

• Poor quality
• Lack of data science expertise
• Non-representative
• Cost of staffing with data scientists
• Insufficient
Data Users • Lack of management support
• Overfitting and
underfitting

• Complexity in formulating questions • Data privacy issues

• Explaining models to the business • Tool selection can be complicated
• Cost of building systems • Integration with other systems
Business Technology

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

40
reserved.
Using existing models and services

• Amazon ML managed
services
Amazon ML
• No ML experience needed
managed services

You Only
• Use existing trained and tuned models
Look Once
• Enhance with domain-specific
(YOLO)
instances
• Over 250 ML model packages and algorithms
• Over 14 industry segments
AWS Marketplace
© 2020, Amazon Web Services, Inc. or its Affiliates. All rights
41
reserved.
 • Machine learning challenges
Section 5 key • Data
takeaways • People
• Business
• Technology
• Managed services simplify
machine learning

42 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module 2: Introduction to Machine Learning

Module wrap-up

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

reserved.
 • Machine learning is a subset of artificial
Module intelligence
• Machine learning applies learning algorithms to
takeaways develop models from large datasets
• The machine learning pipeline describes the
different stages for developing a machine
learning application
• The Amazon Machine Learning stack has three
key layers
• Managed services, machine learning services,
machine learning frameworks
• Machine learning development is different from
traditional development
• Training algorithm is applied to data to create a
model for making predictions

44 © 2020, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Module summary

In summary, in this module, you learned how to:

• Recognize how machine learning and deep learning are part of artificial
intelligence
• Describe artificial intelligence and machine learning terminology
• Identify how machine learning can be used to solve a business problem
• Describe the machine learning process
• List the tools available to data scientists
• Identify when to use machine learning instead of traditional software
development methods

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

45
reserved.
Complete the knowledge check

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

46
reserved.
Additional resources

• What is Machine Learning?

• Machine Learning on AWS

© 2020, Amazon Web Services, Inc. or its Affiliates. All rights

47
reserved.
Thank you

© 2021 Amazon Web Services, Inc. or its affiliates. All rights reserved. This work may not be reproduced or redistributed, in whole or in part, without prior written permission
from Amazon Web Services, Inc. Commercial copying, lending, or selling is prohibited. Corrections, feedback, or other questions? Contact us at
https://support.aws.amazon.com/#/contacts/aws-training. All trademarks are the property of their owners.